Predictive modeling of metrology in semiconductor processes

Abstract

Implementations described herein generally relate to improving silicon wafer manufacturing. In one implementation, a method includes receiving data from one or more manufacturing tools about a manufacturing process of a silicon wafer. The method further includes determining, based on the data, predictive information about a quality of the silicon wafer. The method further includes providing the predictive information to a manufacturing system, wherein the predictive information is used to determine whether to take corrective action.

The invention claimed is: 
     
       1. A method for improving silicon wafer manufacturing, comprising:
 receiving time-series data from one or more manufacturing tools about a manufacturing process of a silicon wafer; 
 converting the time-series data from a parameter-specific and measurement unit-specific space comprising a first number of dimensions into a parameter-agnostic and measurement unit-agnostic space comprising a second number of dimensions to produce converted data; 
 determining, based on the converted data, predictive information about a quality of the silicon wafer; 
 providing the predictive information to a manufacturing system, wherein the predictive information is used to determine whether to take corrective action. 
 
     
     
       2. The method of  claim 1 , wherein the predictive information is determined using stochastic modeling. 
     
     
       3. The method of  claim 1 , wherein the time-series data comprises information about behavior of one or more parameters. 
     
     
       4. The method of  claim 1 , wherein the predictive information comprises information about at least one of: a thickness, an optical reflective index, an absorption index, a strength, and a critical dimension variation across the silicon wafer. 
     
     
       5. The method of  claim 1 , further comprising:
 providing the predictive information to a metrology tool, wherein the metrology tool checks for a problem based on the predictive information. 
 
     
     
       6. The method of  claim 1 , wherein determining the predictive information further comprises converting a result of one or more calculations back into the parameter-specific and measurement unit-specific space. 
     
     
       7. A computing system, comprising:
 a memory; and 
 a processor configured to perform a method for improving silicon wafer manufacturing, the method comprising: 
 receiving time-series data from one or more manufacturing tools about a manufacturing process of a silicon wafer; 
 converting the time-series data from a parameter-specific and measurement unit-specific space comprising a first number of dimensions into a parameter-agnostic and measurement unit-agnostic space comprising a second number of dimensions to produce converted data; 
 determining, based on the converted data, predictive information about a quality of the silicon wafer; 
 providing the predictive information to a manufacturing system, wherein the predictive information is used to determine whether to take corrective action. 
 
     
     
       8. The computing system of  claim 7 , wherein the predictive information is determined using stochastic modeling. 
     
     
       9. The computing system of  claim 7 , wherein the time-series data comprises information about behavior of one or more parameters. 
     
     
       10. The computing system of  claim 7 , wherein the predictive information comprises information about at least one of: a thickness, an optical reflective index, an absorption index, a strength, and a critical dimension variation across the silicon wafer. 
     
     
       11. The computing system of  claim 7 , wherein the method further comprises:
 providing the predictive information to a metrology tool, wherein the metrology tool checks for a problem based on the predictive information. 
 
     
     
       12. The computing system of  claim 7 , wherein determining the predictive information further comprises converting a result of one or more calculations back into the parameter-specific and measurement unit-specific space. 
     
     
       13. A non-transitory computer-readable medium comprising instructions that when executed by a computing device cause the computing device to perform a method for improving silicon wafer manufacturing, the method comprising:
 receiving time-series data from one or more manufacturing tools about a manufacturing process of a silicon wafer; 
 converting the time-series data from a parameter-specific and measurement unit-specific space comprising a first number of dimensions into a parameter-agnostic and measurement unit-agnostic space comprising a second number of dimensions to produce converted data; 
 determining, based on the converted data, predictive information about a quality of the silicon wafer; 
 providing the predictive information to a manufacturing system, wherein the predictive information is used to determine whether to take corrective action. 
 
     
     
       14. The non-transitory computer-readable medium of  claim 13 , wherein the predictive information is determined using stochastic modeling. 
     
     
       15. The non-transitory computer-readable medium of  claim 13 , wherein the time-series data comprises information about behavior of one or more parameters. 
     
     
       16. The non-transitory computer-readable medium of  claim 13 , wherein the predictive information comprises information about at least one of: a thickness, an optical reflective index, an absorption index, a strength, and a critical dimension variation across the silicon wafer. 
     
     
       17. The non-transitory computer-readable medium of  claim 13 , wherein the method further comprises:
 providing the predictive information to a metrology tool, wherein the metrology tool checks for a problem based on the predictive information. 
 
     
     
       18. The non-transitory computer-readable medium of  claim 13 , wherein determining the predictive information further comprises converting a result of one or more calculations back into the parameter-specific and measurement unit-specific space.